multi-hop-rag

Name: multi-hop-rag
Availability: InStock
Author: Zilliz

Decompose questions, retrieve across documents, and synthesize cited answers.

3|2|Updated Jan 29, 2026

Search Tags:#llm #rag #retrieval #vector-search #multihop #milvus

Ready to run

npx skills add https://github.com/zilliztech/milvus-marketplace --skill multi-hop-rag

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: multi-hop-rag
Source: https://github.com/zilliztech/milvus-marketplace/tree/main/plugins/rag-toolkit/skills/multi-hop-rag
Command: npx skills add https://github.com/zilliztech/milvus-marketplace --skill multi-hop-rag

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Solves complex questions that require information from multiple documents by chaining retrievals and synthesizing evidence across sources.

Core Features & Use Cases

Cross-document reasoning: connect facts across sources to form a coherent answer.
Iterative retrieval: perform multiple hops, refining results between steps.
Transparent synthesis: present a synthesized answer with cited sources and context.

Quick Start

Example: Ask a complex question such as "Which documents mention X and how do they relate to Y?" The system will decompose the question, run hops across sources, and return a synthesized answer with sources.

Frequently Asked Questions about multi-hop-rag

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I answer complex questions that need information from multiple documents?▼

Multi-hop retrieval chains retrievals across documents to synthesize evidence. It decomposes your question into steps, retrieves relevant passages iteratively, and synthesizes a coherent answer with cited sources—essential for cross-document reasoning in research, legal, and enterprise contexts.

What's the difference between standard RAG and multi-hop retrieval?▼

Standard RAG retrieves once; multi-hop RAG performs iterative retrievals, refining results between steps to connect facts across sources. This approach handles complex questions requiring reasoning across multiple documents that single-pass retrieval cannot answer completely.

What do I need to set up before using multi-hop retrieval?▼

You need a vector store for document embeddings, an LLM for reasoning and synthesis, and orchestration logic to decompose questions and chain retrieval steps. The vector store indexes your documents; the LLM manages multi-step reasoning.

Can multi-hop retrieval work with vector search and Milvus?▼

Yes. Multi-hop RAG uses vector search to retrieve passages across hops and integrates with vector stores like Milvus for scalable document indexing and retrieval across multiple steps.

How do I trace how multi-hop retrieval found an answer?▼

Multi-hop RAG returns synthesized answers with cited sources and context from each retrieval hop, making the reasoning path transparent. You can see which documents contributed to each step and how facts connected across sources.

When should I use multi-hop retrieval instead of simpler approaches?▼

Use multi-hop retrieval when questions require connecting facts across multiple documents, iterative discovery, or reasoning that single-pass retrieval cannot handle. It's suited for complex research, legal analysis, and enterprise knowledge synthesis.